summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* First attempt at modelling leakageexp_leakage4Jean-Marc Valin2017-05-313-9/+72
|
* silk_LIMIT_32() should return an opus_int32 (not opus_int)Jean-Marc Valin2017-05-301-1/+1
| | | | | Thanks to petrufm for pointing that out: https://github.com/xiph/opus/issues/35
* Fix memory leaks in opus_compare.cJean-Marc Valin2017-05-261-0/+3
| | | | This makes it possible to run the testvectors with asan enabled.
* Fixes unit tests that need arch-specific codeJean-Marc Valin2017-05-259-152/+69
| | | | | We now include the object files for those rather than attempt to problems.
* appveyor: fix typoed appveyor commandRicardo Constantino2017-05-251-1/+1
| | | | | | | And run the build request only at the last configuration. Also add a check if api_key is unset to not try to request a new build. Signed-off-by: Ralph Giles <giles@thaumas.net>
* appveyor: Add the api key for the linked ci builds.Ralph Giles2017-05-241-1/+1
| | | | | | | | | The appveyor script needs an encrypted api key to trigger downstream builds, and there doesn't seem to be an easy way to make the builds we run for the main repo something other contributors can control. Signed-off-by: Ricardo Constantino <wiiaboo@gmail.com>
* appveyor: request rebuild of opus-tools on successful buildRicardo Constantino2017-05-241-0/+7
| | | | | | Using my API key for now, but needs rillian's to work. Signed-off-by: Ralph Giles <giles@thaumas.net>
* appveyor: simplify and preserve pathsRicardo Constantino2017-05-241-18/+6
| | | | | | Artifact with preserved paths is more useful for CI. Signed-off-by: Ralph Giles <giles@thaumas.net>
* appveyor: move to dotfile like travisRicardo Constantino2017-05-241-0/+0
| | | | | | Less clutter on systems that hide dotfiles. Signed-off-by: Ralph Giles <giles@thaumas.net>
* Updated SHA256 of releasesJean-Marc Valin2017-05-241-0/+6
|
* Fixes a double->float conversion warningJean-Marc Valin2017-05-241-1/+1
|
* Fixes some fixed-point 16-bit int overflowsv1.2-betaJean-Marc Valin2017-05-242-2/+2
| | | | | The code would have run fine on 32-bit archs, but would have overflowed on a 16-bit arch
* Reducing trim at low bitrateJean-Marc Valin2017-05-241-2/+10
| | | | | | | | | | | | Some informal tests seem to confirm that reducing the trim at 32-64 kbps improves quality (better HF). It's not clear whether it's also the case at 96 kb/s and above, so we're leaving it as is for those rates. This corresponds to buildC in this thread: https://hydrogenaud.io/index.php/topic,113985.0.html Also see: https://hydrogenaud.io/index.php/topic,111798.0.html
* cosmetics,silk: correct input/output arg commentsLinfeng Zhang2017-05-238-17/+17
| | | | | | Change-Id: I607a8b75b0711a485384d6f854cf6e2ec18b38b7 Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Fixes a surround bug on loud signalJean-Marc Valin2017-05-231-1/+1
| | | | | The constant was copied from the analysis code which uses +/-1 as float scaling, but surround_analysis() uses +/-32767.
* Optimize silk_biquad_alt_stride2() for ARM NEONLinfeng Zhang2017-05-178-5/+251
| | | | | | | | The optimization is bit exact with C function. Change-Id: Ifb8f04b19f2d576e79ce5dcfa7e0fc374d71d6c8 Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Update silk_biquad_alt()Linfeng Zhang2017-05-174-12/+64
| | | | | | | | | | | Split to silk_biquad_alt_stride1() and silk_biquad_alt_stride2(), so that it can be optimized more efficiently when stride is 2. This change in C code is bit exact with the origin. Change-Id: Idaefe670397016ace2a489e3435ac61b7dbe79d5 Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Indent commands in the README.Ralph Giles2017-05-101-16/+17
| | | | | | Makes things a little easier to read. Signed-off-by: Gian-Carlo Pascutto <gcp@sjeng.org>
* Suggest basic build dependencies in the README.Ralph Giles2017-05-101-0/+18
| | | | | | | Make it easier for users unfamiliar with C applications to installed the necessary build dependencies. Signed-off-by: Gian-Carlo Pascutto <gcp@sjeng.org>
* Avoid warning when __STDC_VERSION__ is not definedCarlos Alberto Lopez Perez2017-05-011-1/+1
| | | | | | | | This fixes an issue with gcc 4.9 on Debian 8, at least, which defines __STDC__ but not __STDC_VERSION__, unlike more recent gcc. Signed-off-by: Ralph Giles <giles@thaumas.net>
* oops, fix build broken by previous patchJean-Marc Valin2017-04-281-2/+0
|
* Non-diegetic support for Ambisonics Mapping 254.Drew Allen2017-04-281-23/+90
| | | | Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Optimize silk_warped_autocorrelation_FIX() for ARM NEONLinfeng Zhang2017-04-1311-11/+376
| | | | | | | | | | | The optimization is bit exact with C function. This optimization speeds up fixed-point SILK encoder on NEON about 5% to 8%. (Tested on Acer Chromebook, ARMv7 Processor rev 3 (v7l).) Change-Id: I582f6f3585b7946149e16a2ad3084ebc0ae79a4f Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Fix OPUS_ARG_NONNULL indices in opus_multistream.hFelicia Lim2017-03-271-2/+2
|
* VS2015: remove previously ignored C4146 in kiss_fft.cRicardo Constantino2017-03-021-3/+1
| | | | | | | | C4146: unary minus operator applied to unsigned type, result still unsigned Fixed in a previous commit. Signed-off-by: Mark Harris <mark.hsj@gmail.com>
* VS2015: Ignore LNK4221 and properly force SSE off for Win32Ricardo Constantino2017-03-021-0/+4
| | | | | | | | | This warning complains of files that don't define any new symbols not being included in linking. GCC doesn't seem to complain about those, so neither should VS. Signed-off-by: Mark Harris <mark.hsj@gmail.com> Closes https://github.com/xiph/opus/pull/34
* VS2015: ignore C4244 warning in opus_compare.cRicardo Constantino2017-03-021-1/+3
| | | | | | | | opus_compare.c defines the bitstream so we avoid changing it by ignoring the non-breaking warning instead of changing the file. Signed-off-by: Mark Harris <mark.hsj@gmail.com> Resolves https://github.com/xiph/opus/issues/21
* Fix compiler warningsMark Harris2017-02-2622-90/+91
| | | | | | | | | | | | | | | | | | | - celt/modes.c:430:14: warning: cast from 'const unsigned char *' to 'opus_int16 *' increases required alignment from 1 to 2 [-Wcast-align] - 'C[0][1]' may be used uninitialized [-Wmaybe-uninitialized] - Unused variable/parameter - Value stored is never read - MSVC warnings about "possible loss of data" due to type conversions - MSVC warning C4146: unary minus operator applied to unsigned type - silk/NLSF_del_dec_quant.c:137:20: warning: array subscript is above array bounds [-Warray-bounds] (gcc -O3 false positive) - src/mlp_train.h:39:20: warning: function declaration isn't a prototype [-Wstrict-prototypes] - Remove SMALL_FOOTPRINT code from SSE 4.1 FIR implementation, matching the C implementation. The clang -Wcast-align warnings with SSE intrinsics are a known clang issue: https://llvm.org/bugs/show_bug.cgi?id=20670
* Add common.props to Makefile.amRicardo Constantino2017-02-201-0/+1
| | | | Signed-off-by: Mark Harris <mark.hsj@gmail.com>
* VS2015: only compile fixed silk for appropriate configsRicardo Constantino2017-02-202-324/+28
| | | | | Signed-off-by: Mark Harris <mark.hsj@gmail.com> Closes https://github.com/xiph/opus/pull/26
* VS2015: Move back PreprocessorDefinitions to opusRicardo Constantino2017-02-202-2/+2
| | | | | | They're only needed in that project, no need to be in common.props. Signed-off-by: Mark Harris <mark.hsj@gmail.com>
* VS2015: join common options in a properties fileRicardo Constantino2017-02-206-2903/+204
| | | | | | | | | | | | | | Should make the projects a bit easier to digest and the differences between the different configurations and platforms be more obvious. A (happy?) side effect of this is the static libraries are now about the same size as the ones produced by GCC and seem to build fine with opus-tools. Needs testing with other third party applications to make sure nothing is missing that shouldn't. Signed-off-by: Mark Harris <mark.hsj@gmail.com>
* VS2015: remove dead options for linker in static configurationsRicardo Constantino2017-02-201-19/+2
| | | | Signed-off-by: Mark Harris <mark.hsj@gmail.com>
* test_opus_decode: Fix build on GCC 3.4 to 4.5.xMark Harris2017-02-181-2/+2
| | | | Broken by b0949f11, reported by ko-zu on GitHub.
* Fix sum initializer in celt_fir()Jean-Marc Valin2017-02-181-4/+5
|
* Fix stability testJean-Marc Valin2017-02-181-1/+1
|
* Speed up test_unit_LPC_inv_pred_gain by returning early on stable filtersJean-Marc Valin2017-02-181-0/+10
|
* Remove SMALL_FOOTPRINT code for celt_fir()Jean-Marc Valin2017-02-171-20/+10
| | | | | Wasn't worth it given the small code size of the alternative, which also got refactored a little (still bit exact).
* Increasing GLOBAL_STACK_SIZE to 120000 to avoid failureJean-Marc Valin2017-02-151-2/+2
| | | | | | Cannot prove it's the correct value, but it's better than the previous values, which sometimes segfaults. The increase was made necessary due to 120 ms frame size support.
* Reduce the scope of Ne10 includesMichael Bradshaw2017-02-155-5/+4
| | | | | | | libopus only uses the DSP module of Ne10, and never uses the init functions. Signed-off-by: Michael Bradshaw <mjbshaw@google.com> Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Remove #include "config.h" from ARM Ne10 headersMichael Bradshaw2017-02-152-2/+0
| | | | | Signed-off-by: Michael Bradshaw <mjbshaw@google.com> Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Optimize silk_LPC_inverse_pred_gain() for ARM NEONLinfeng Zhang2017-02-1420-21/+386
| | | | | | | | The optimization is bit exact with C function. Change-Id: Ib3bdc26a5a4ebe02e7f24be85104e8e9a2a9a738 Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Adding a unit test for LPC_inv_pred_gain()Jean-Marc Valin2017-02-143-4/+156
| | | | | | | It checks that no clearly unstable filter passes the LPC_inv_pred_gain() test. Also, this will make it possible to check assembly for correctness. Modified from an original patch from Linfeng Zhang <linfengz@google.com>.
* Add more coverage to travis tests.Stephen2017-02-141-0/+8
| | | | Signed-off-by: Ralph Giles <giles@mozilla.com>
* Rename y0/y1 to out0/out1 to avoid name clash with the Bessel functionsJean-Marc Valin2017-02-141-5/+5
|
* Add decoder fuzz target.Felicia Lim2017-02-102-0/+127
|
* Remove silk_LPC_inverse_pred_gain_Q24() which is no longer used anywhereJean-Marc Valin2017-02-092-26/+0
|
* arm2gnu.pl: Fix GNU ARM .type directiveMark Harris2017-02-041-2/+2
| | | | | | Broken by 76e831d. Without the .type directive, SIGILL may be produced if the C code is compiled in Thumb mode, because the compiler may assume that the asm symbol is also Thumb and call it using a BL instruction.
* fix ARM build w/--disable-intrinsics --enable-asmJames Zern2017-02-041-1/+1
| | | | | | | | | and rtcd disabled (CFLAGS=-mfpu=neon) broken since: cfdaf365 Optimize silk_NSQ_del_dec() for ARM NEON Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Optimize silk_NSQ_del_dec() for ARM NEONLinfeng Zhang2017-01-307-1/+1257
| | | | | | | | | | | | | | | | The optimization is bit exact with C function. This optimization speeds up SILK encoder on NEON as following. Fixed-point: Complexity 0-5: 0% Complexity 6-7: 6% Complexity 8-9: 10% Complexity 10: 8% Got similar results on floating-point. Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
View Lorry Controller Status